Lot Production Anomaly Detection with Bathtub Kernel Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manufacturing systems face challenges in detecting anomalies in product characteristics during the production process, leading to potential defects and inefficiencies, especially in complex processes like semiconductor device fabrication where anomalies may not be identified until the final testing stage, causing significant quality loss and delivery issues.
Innovation Solution
Implementing a Gaussian process regression model with a bathtub kernel function to detect anomalies in product characteristics using data from upstream quality control processes, allowing for predictive distribution generation and real-time adjustment of manufacturing settings to prevent defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional quality control procedures are used throughout the manufacturing process, then product quality can be monitored, but anomalies are not detected until the final testing stage causing significant quality loss
Solution Approach 1:
The patent applies preliminary action by using a Gaussian process regression model to predict product characteristics at intermediate quality control steps before the final testing stage. This allows anomalies to be detected early in the manufacturing process, enabling corrective actions to be taken before defects are confirmed, thus reducing both quality loss and detection time.
2Productivity
If predictive methods are used to reduce defects, then manufacturing efficiency improves, but the system complexity increases
Solution Approach 1:
The patent uses an intermediary approach by introducing a Gaussian process regression model as a predictive layer between the quality control steps and the final product evaluation. This model acts as a mediator that processes quality control data and predicts final product characteristics, enabling early anomaly detection without requiring complete system redesign, thus balancing productivity improvement with manageable complexity.
3Measurement precision
If Gaussian process regression model with bathtub kernel function is implemented, then anomaly detection accuracy improves, but computational requirements increase
Solution Approach 1:
The patent applies parameter changes by utilizing a bathtub kernel function with specific parameters (α, β, γ, δ) that control the shape and characteristics of the kernel. These parameter adjustments optimize the balance between anomaly detection accuracy and computational efficiency, allowing the Gaussian process regression model to achieve high precision while managing computational resource requirements through carefully tuned kernel parameters.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure describes a computer-implemented method for detecting anomalies during lot production, wherein the products within a production lot are processed according to a sequence of steps that include manufacturing steps and one or more quality control steps interspersed among the manufacturing steps, the method comprising: obtaining process quality inspection data from each of the one or more quality control steps for a first production lot; obtaining product characteristics data for the products in the first production lot after the final step in the sequence; training a Gaussian process regression model using the process quality inspection data and the product characteristics data from the first production lot; generating a predictive distribution of the product characteristics data using the Gaussian process regression model that uses a bathtub kernel function; obtaining process quality inspection data from each of the quality control steps for a second production lot; identifying anomalies in the second production lot using the predictive distribution of the product characteristics data and the process quality inspection data from the second production lot; if no anomalies are detected in the second production lot, updating the Gaussian process regression model using the process quality inspection data from the second production lot; setting target values for one or more values in the process quality inspection data based on the predictive distribution of the product characteristic; and adjusting settings of one or more manufacturing steps based on the target values.